Ketene-Ketene interconversion. 6-Carbonylcyclohexa-2,4-dienone-hepta-1,2,4, 6-tetraene-1,7-dione-6-oxocyclohexa-2,4-dienylidene and wolff rearrangement to fulven-6-one

Article abstract of DOI:10.1021/jo5011087

6-Carbonylcyclohexa-2,4-dienone (1) has been generated by flash vacuum thermolysis (FVT) with Ar-matrix isolation of methyl salicylate (7), 2-phenylbenzo-1,3-dioxan-4-one (8), phthalic peranhydride (9), and benzofuran-2,3-dione (11) and also by matrix photolysis of 9, 11, and 2-diazocyclohepta-4,6-dien-1,3-dione (12). In each case, FVT above 600 °C results in decarbonylation of 1 and Wolff rearrangement to fulven-6-one (13) either concertedly or via open-shell singlet 6-oxocyclohexa-2,4-dienylidene (18). Ketenes 1 and 13 were characterized by IR spectroscopy. Photolysis of matrix-isolated 1 at 254 nm also results in the slow formation of 13. The sequential formation of ketenes 1 and 13 from 7 has also been monitored by FVT-mass spectrometry, and 13 has been trapped with MeOH to afford methyl 1,3-cyclopentadiene-1- and -2-carboxylates 15 and 16. FVT of methyl salicylate-1-¹³C 7a revealed a deep-seated rearrangement of the ¹³C-labeled 1a to hepta-1,2,4,6-tetraen-1,7-dione (17a) by means of electrocyclic ring opening followed by a facile 1,5-H shift and recyclization prior to CO-elimination and ring contraction to ¹³C-labeled 13. The rearrangement mechanism is supported by M06-2X/6-311++G(d,p) calculations, which predict feasible barriers for the FVT rearrangements and confirm the observed labeling pattern in the isolated methyl salicylate 7a/7b and methyl cyclopentadienecarboxylates 20 and 21 resulting from trapping of 13 with MeOH.

Full text of DOI:10.1021/jo5011087

Article
pubs.acs.org/joc
Ketene−Ketene Interconversion. 6‑Carbonylcyclohexa-2,4-dienone−
Hepta-1,2,4,6-tetraene-1,7-dione−6-Oxocyclohexa-2,4-dienylidene
and Wolff Rearrangement to Fulven-6-one
Rainer Koch,*,† Rodney J. Blanch,‡ and Curt Wentrup*,‡
†Institut fur Chemie and Center of Interface Science, Carl von Ossietzky Universitat Oldenburg, P.O. Box 2503, 26111 Oldenburg,
̈
̈
Germany
‡School of Chemistry and Molecular Biosciences, The University of Queensland, Brisbane, Qld 4072, Australia
S
* Supporting Information
ABSTRACT: 6-Carbonylcyclohexa-2,4-dienone (1) has been gener-
ated by flash vacuum thermolysis (FVT) with Ar-matrix isolation of
methyl salicylate (7), 2-phenylbenzo-1,3-dioxan-4-one (8), phthalic
peranhydride (9), and benzofuran-2,3-dione (11) and also by matrix
photolysis of 9, 11, and 2-diazocyclohepta-4,6-dien-1,3-dione (12). In
each case, FVT above 600 °C results in decarbonylation of 1 and Wolff
rearrangement to fulven-6-one (13) either concertedly or via open-shell
singlet 6-oxocyclohexa-2,4-dienylidene (18). Ketenes 1 and 13 were
characterized by IR spectroscopy. Photolysis of matrix-isolated 1 at 254
nm also results in the slow formation of 13. The sequential formation
of ketenes 1 and 13 from 7 has also been monitored by FVT-mass
spectrometry, and 13 has been trapped with MeOH to afford methyl
1,3-cyclopentadiene-1- and -2-carboxylates 15 and 16. FVT of methyl salicylate-1-13C 7a revealed a deep-seated rearrangement of
the 13C-labeled 1a to hepta-1,2,4,6-tetraen-1,7-dione (17a) by means of electrocyclic ring opening followed by a facile 1,5-H shift
and recyclization prior to CO-elimination and ring contraction to 13C-labeled 13. The rearrangement mechanism is supported by
M06-2X/6-311++G(d,p) calculations, which predict feasible barriers for the FVT rearrangements and confirm the observed
labeling pattern in the isolated methyl salicylate 7a/7b and methyl cyclopentadienecarboxylates 20 and 21 resulting from
trapping of 13 with MeOH.
Scheme 1. Gompper’s Proposal
INTRODUCTION
■
Ketenes, RR′CCO, are usually thermodynamically stable
but kinetically highly reactive compounds useful in cyclo-
addition and electrocyclization reactions and nucleophilic
addition chemistry.1 Several rearrangements involving ketenes
are known, e.g., the Wolff2 and retro-Wolff3 rearrangements,
the α-oxoketene−α-oxoketene rearrangement, and similar
rearrangements of imidoylketenes, acylthioketenes, and vinyl-
ketenes.4 Ketenes can eliminate CO on flash vacuum
thermolysis (FVT), thereby generating carbenes, which may
undergo further rearrangements.5
In an investigation of the mechanism of benzyne formation
by thermolysis of benzenediazonium carboxylate (4), Gompper
et al. demonstrated on the basis of kinetic measurements the
intermediacy of species that could be trapped with water and
methanol to generate salicylic acid (6) and methyl salicylate (7)
together with the main products derived from benzyne (5).6
Logically, the formation of 6-oxocyclohexa-2,4-dienylideneke-
tene (1) via benzoxetone (2) and carboxylate 3 was postulated
(Scheme 1), but other work suggests that this is only a minor
route to 1 and 2.7
2).8 Photolysis of an Ar matrix of 9 at λ > 315 nm also gave a
ketene, presumably 1, now absorbing at 2139 and 1650 cm−1.9
Further photolysis of this species at λ > 340 nm converted it
into benzoxet-2-one 2 in a photochemically reversible process,
and continued photolysis yielded benzyne 5.9 Calculations at
the QCISD//MP2 level indicate a very small free energy
difference between 1 and 2, the ketene 1 being more stable by
ca. 2.8 kcal/mol.10 Ketene 1 was trapped with MeOH to afford
The photolysis of 2-phenylbenzo-1,3-dioxin-4-one (8) at 77
Received: May 20, 2014
Published: July 2, 2014
K afforded an IR spectrum ascribed to 1 (2118 cm−1) (Scheme
© 2014 American Chemical Society
6978
dx.doi.org/10.1021/jo5011087 | J. Org. Chem. 2014, 79, 6978−6986

The Journal of Organic Chemistry
Article
and 100 × 107 M−1 s−1, respectively. No barrier was found for
the addition of H2O at the MP2 computational level.10a,16
Gas-phase pyrolysis of methyl salicylate (7) and benzofur-
andione (11) yields fulven-6-one (carbonylcyclopentadiene)
(13). Both ketenes 1 and 13 were observed by photoelectron
spectroscopy,17 but pyrolysis−mass spectrometry failed to
detect ketene 1.18 Our work, detailed below, has provided
clear evidence for the sequential formation of 1 and 13 by FVT
as evidenced by IR and mass spectrometry as well as trapping
with MeOH. Moreover, we report a cascade rearrangement of
α-oxoketene 1 involving electrocyclic ring opening, 1,5-H shift,
CO elimination, and finally Wolff rearrangement to 13. This
new rearrangement mechanism is supported by 13C labeling as
well as computational studies.
Scheme 2. Precursors for α-Oxoketene (1) and
Carbonylcyclopentadiene (13)
RESULTS AND DISCUSSION
■
1. Generation of α-Oxoketene (1). Argon matrices of 1
were prepared from five different precursors. FVT of 7, 8, 9,
and 11 resulted in elimination of MeOH, benzaldehyde, CO2,
and CO, respectively. Ketene 1 was also obtained by matrix
photolysis of 9, 11, and 12. Identical IR spectra of α-oxoketene
(1) were obtained in each case and dominated by the very
strong CCO absorption at 2134 cm−1, which is flanked by
a strong absorption at 2145 cm−1 and a weaker one at 2114
cm−1. Examples are shown in Figure 1 and in Figure S1
(Supporting Information). The observed IR spectra are in very
good agreement with the calculated harmonic vibrational
spectrum at the B3LYP/6-31G(d) level (see Figure S2,
Supporting Information). It is well established that α-
oxoketenes can exhibit multiple IR absorption bands in the
cumulene region.19 When the matrices are slowly warmed to
remove the argon, the multiple absorptions coalesce to a single
peak at 2143 cm−1. The neat ketene is observable at
temperatures up to 150 K.
methyl salicylate (7).9 Formation of 1 by treatment of salicyloyl
chloride with Hunig’s base and by solid-state thermolysis of 10
̈
at 130 °C was also postulated,11 but the reported IR
frequencies of 1 and 2 (2070, 2050, and 1930 cm−1) are not
in good agreement with current values (2134−2145 and 1904
cm−1, respectively; see below). Similarly, the thermolysis of
phenyl salicylate (salol) was postulated to proceed via
oxoketene (1).12 The photolysis of 11 in the presence of
water and carboxylic acids gave salicylic acid 6 and mixed
anhydrides thereof, and the intermediacy of 1 was postulated.13
A UV spectrum ascribed to 1 was obtained on photolysis of 8,
9, and 10, but IR spectra were not observed.14
The IR spectrum of 1 in Ar matrix was obtained by flash
vacuum thermolysis (FVT) of methyl salicylate 7 in 1994, and
the same IR spectrum was obtained by FVT of 9 and 11 and
also by matrix photolysis of 9, 11, and 12.15 Four years later, an
IR absorption at 2135 cm−1 was ascribed to 1 in acetonitrile
solution by means of time-resolved IR spectroscopy.10a This
study permitted the evaluation of the rates of reaction of 1 with
H2O, MeOH, and Et2NH at room temperature as ca. 1.5, 3.0,
In each case, the photochemical interconversion9a of α-
oxoketene (1) and benzoxetone (2) was used as a diagnostic
tool to confirm the correct assignment of spectra due to 1. An
example is shown in Figure 2.
Figure 1. IR difference spectrum resulting from irradiation at λ > 360 nm of (top) diazocycloheptadienedione (12); (bottom) benzofurandione 11.
A = ketene 1; B = benzoxetone 2; C = 12; D = 11 (products are shown as positive peaks and precursors as negative peaks in both spectra). All
spectra in Ar matrix at 10 K.
6979
dx.doi.org/10.1021/jo5011087 | J. Org. Chem. 2014, 79, 6978−6986

The Journal of Organic Chemistry
Article
Figure 2. IR difference spectrum showing ketene 1 (A) formed by FVT of methyl salicylate (7) at 800 °C and condensed in Ar matrix at 10 K
(negative peaks). Positive peaks are due to benzoxetone 2 (B) formed on irradiation of the matrix at λ > 340 nm. A: 2145 (m), 2134 (vs) and 2114
(w), 1650, 1521, 1292 cm−1. B: 1916, 1904, 1624, 1604, 1443, 1340 cm−1. Inset: expansion of the ketene peaks A (2145, 2134, 2114 cm−1).
2. Generation of Carbonylcyclopentadiene (13). The
formation of α-oxoketene (1) starts at 300 °C in the FVT of
methyl salicylate (7) and reaches a maximum at 650−700 °C.
At temperatures around 700 °C, a new ketene starts appearing,
which absorbs at 2135 (vvs) and 2131 (vs) cm−1 in the Ar
matrix. This compound was identified as fulven-6-one (carbon-
ylcyclopentadiene) (13) by comparison with an authentic
sample prepared by matrix photolysis of 2-diazo-3,5-cyclo-
hexadienone 1420 (Scheme 3 and Figures S3 and S4,
agreement with this, a trace of naphthalene was isolated from
the preparative FVT reactions, but not seen in the matrix
isolation experiments, where the lower pressure and presence of
Ar gas would disfavor dimerization. Naphthalene is the known
thermal rearrangement product of the dimer of cyclo-
pentadienylidene, fulvalene.20a,18
Thermal elimination of CO from ketenes has been observed
in several other instances.5 Thus, high-temperature elimination
of CO from ketenes 1 and 13 is not surprising. In the case of
13, CO elimination will be facilitated by the fact that this
ketene is born with an excess energy of more than 60 kcal/mol
(see the Computational Details).
Scheme 3. Formation of Carbonylcyclopentadiene (13)
In summary, and in contrast to the unsuccessful FVT-MS
experiments of Mamer et al.,18 the sequential formation of 1
and 13 on FVT of 7 is clearly established by the IR and MS
investigations reported here (Scheme 4).
Supporting Information) and with the calculated vibrational
spectrum (Figure S2, Supporting Information). Other examples
of Wolff-type ring contraction to fulvenones on FVT of oxet-2-
ones, thiet-2-ones, and azet-2-ones have been reported.21
Moreover, photolysis of the matrix-isolated ketene 1 at 254
nm also resulted in decarbonylation and formation of 13. After
gentle warming of matrices of 13 to allow the Ar to evaporate,
this ketene was observable at temperatures up to 180 K. This is
important, because it will allow us to isolate 13 on a coldfinger
at 77 K in order to trap it with MeOH as described below.
The thermal decarbonylation of 1 with formation of 13 was
monitored by online mass spectrometry.22 The molecular ion
of methyl salicylate 7•+ (m/z 152) disappears at FVT
temperatures above 700 °C (Figure 3). The ion corresponding
to 1•+ (m/z 120) is both a fragment peak in the mass spectrum
of 723 and a thermal reaction product. This ion, 1•+ (m/z 120),
reaches maximum intensity at ca. 680 °C (Figure 4) when 7•+
has almost disappeared (Figure 4). At higher temperatures, m/z
120 decreases rapidly, as fulven-6-one 13 is formed thermally
above ca. 700 °C (m/z 92, Figure 4). All the while, m/z 64
(formally cyclopentadienylidene) keeps increasing because it is
a fragment ion of both 7•+ and 13•+, but especially of 13•+. A
further increase in the intensity of m/z 64 at the highest
temperature suggests that thermal decarbonylation of ketene 13
to form cyclopentadienylidene may be taking place. In
3. Trapping of the Ketenes with MeOH. Preparative
FVT reactions22 of methyl salicylate were carried out with
trapping of the product on a liquid N2-cooled coldfinger (77
K). The yields of products were determined by GC−MS. When
the FVT reaction was carried out at 600 °C, subsequent
warming of the coldfinger afforded methyl salicylate 7 (92%)
together with cyclopentadienecarboxylates 15 and 16 (4%) as a
result of trapping of ketenes 1 and 13 with methanol (Scheme
5). The corresponding FVT reaction at 780 °C afforded methyl
salicylate 7 (52%) and cyclopentadienecarboxylates 15 and 16
(41%) in accord with the thermal processes described in
Sections 1 and 2 (Figure S5, Supporting Information). The
combined yields of these products were increased to 100%
when excess methanol was present. Two methods were used for
trapping with excess methanol: (i) A layer of MeOH is
deposited on the coldfinger prior to the FVT reaction. The
FVT product is deposited on top of this layer, and finally a new
layer of MeOH is deposited on top of the pyrolyzate. This
methanol matrix melts on warming and is allowed to flow into a
receiving flask. (ii) The gaseous thermolysis product can be
mixed with MeOH vapor, which is passed through a concentric
tube terminating shortly before the exit of the FVT tube. The
FVT product is thus collected in a methanol matrix on the
coldfinger.22,24
6980
dx.doi.org/10.1021/jo5011087 | J. Org. Chem. 2014, 79, 6978−6986

The Journal of Organic Chemistry
Article
Figure 3. EI mass spectra of the products of online FVT of methyl salicylate 7 (m/z 152) at various temperatures: (a) 200 °C; (b) 500 °C; (c) 600
°C; (d) 800 °C. 1 = m/z 120. 13 = m/z 92.
Figure 4. Ion intensities as ion current (IC) versus total ion current (TIC) in the online FVT of 7 (m/z 152) at various temperatures. 1 = m/z 120.
13 = m/z 92. C5H4 = m/z 64.
Scheme 4
atom -% 13C) was synthesized from 1-13C-benzoic acid as
indicated in Scheme 6.
FVT of compound 7a at 800 °C with trapping of the ketene
products with MeOH on the coldfinger afforded a mixture of
recovered but rearranged 13C-labeled methyl salicylate (7b)
together with a mixture of 13C-labeled methyl cyclopentadie-
necarboxylates 20 and 21 (ratio 7:5:1). After thawing of the
MeOH matrix, the sample was kept at −40 °C in order to avoid
4. FVT of Methyl 1-13C-Salicylate 15 Establishing
Ketene Rearrangements. Methyl 1-13C-salicylate (7a) (95
6981
dx.doi.org/10.1021/jo5011087 | J. Org. Chem. 2014, 79, 6978−6986

The Journal of Organic Chemistry
Article
a
Scheme 5. Trapping of Ketenes 1 and 13
Scheme 7. Interpretation of the 13C-Labeling Results
Scheme 6. Synthesis of Methyl 1-13C-Salicylate 7a
a
Note: the thermal interconversions of 1a and 1b with 13C-labeled
benzoxet-2-one 2 are omitted for clarity.
the ketene region around 2100 cm−1 due to 1, 13, and CO
(Figures 1, 2, and S1, Supporting Information) make it
impossible to identify the presence of potential absorptions due
to 17 in this region. However, an otherwise unassigned weak
absorption at 1550 cm−1 seen in some spectra resulting from
the photolyses of 11 and 12 may be due to 17. This peak
disappears on further photolysis (see Figures S10 and S11,
Supporting Information).
The cyclopentadienecarboxylates dimerize at room temper-
ature. For further proof of the labeling pattern, the labeled
dimer 22 was isolated and examined by 13C NMR spectros-
copy, which clearly shows that only the carbon atoms expected
from Scheme 7 are labeled (see Scheme 8 and Figure S12,
Supporting Information). The structure of 22 and the
assignment of the 13C NMR data have been established
unequivocally.25
dimerization of the cyclopentadienes, and NMR spectroscopy
was performed at −40 °C. The composition of the mixture and
the distribution of the 13C label were determined by
comparison of the 13C NMR spectrum with those of the
previously analyzed unlabeled compounds (Figures S7−S9,
Supporting Information). The surprising outcome was that the
label in the methyl salicylate (7b) was now distributed evenly
between carbons 1 and 3 (47% each; see Scheme 7) with only
natural abundance of 13C on all other carbon atoms. In the
methyl cyclopentadiene-1-carboxylate (20) 47% of the label
resided at the ipso carbon C1, and the remaining 47% was
distributed almost evenly between C2 and C5 as indicated in
Scheme 7. Only a natural abundance of 13C was found at all
other carbon atoms. Similar results were obtained for the minor
isomer 21 (Scheme 7 and Figure S9, Supporting Information).
The label distribution in 7b, 20, and 21 demonstrates that a
profound but specific rearrangement is taking place. We explain
this in terms of an electrocyclic ring opening in the initial 13C-
labeled α-oxoketene (1a), which generates the bis-ketene 17a
(Scheme 7). Recyclization will now generate equal amounts of
1a and 1b, which are trapped with MeOH to give the observed
compound 7b. Loss of CO from the equilibrated ketenes 1a
and 1b with Wolff-type rearrangement (concerted or stepwise
via carbene 18) affords the labeled ketene 13a and hence the
isolated cyclopentadienecarboxylates 20 and 21.
Scheme 8. Dimerization of the 13C-Labeled Methyl
Cyclopentadienecarboxylates to 22 (Only One Enantiomer
of the Racemate Is Shown)
The ring-opened ketene 17 is a high energy species lying ca.
44 kcal/mol above ketene 1, and it has only a very small barrier
for the electrocyclization back to 1 and a ca. 10 kcal/mol barrier
for a 1,5-hydrogen shift (see the Calculations). Therefore, the
experimental characterization of 17 is not straightforward.
Ketene 17 is predicted to absorb strongly at 2147, 2105, and
1546 cm−1 (see Figure S2, Supporting Information; wave-
numbers scaled by 0.9614), but the complex IR absorptions in
It is interesting to note that benzoxirene (23) is not involved
in the rearrangements. Had it been, then labeling of the
carboxylate carbon would have occurred, and this was not
detectable (Scheme 9). This is in contrast to the case of the
corresponding iminocarbene, where 13C-labeling demonstrated
the interconversion with 1H-benzazirene prior to ring
contraction to cyclopentadienecarbonitrile.26
6982
dx.doi.org/10.1021/jo5011087 | J. Org. Chem. 2014, 79, 6978−6986

The Journal of Organic Chemistry
Article
required for its formation. The recyclization of 17 to 1 and the
degenerate 1,5-H shift (17 ⇄ 17) have calculated barriers of 3
and 10 kcal/mol, respectively. The triplet ground state (T0) of
the α-oxocarbene, 2-oxocyclohexa-3,5-dienylidene (18), is
predicted to lie 50 kcal/mol above 1. A singlet state of 18
could not be optimized at the present DFT level, not
surprisingly, because the lowest singlet state (S1) is an open
shell, as was also found for the isomeric 4-oxocyclohexadieny-
lidene.28 Calculations on 18 at the CASSCF(8,8)/6-311+
+G(d,p) level afforded a singlet−triplet splitting (S1−T0) of 10
kcal/mol, with the closed-shell singlet S2 lying 21 kcal/mol
above T0 (Scheme 10). The corresponding numbers at the
CASSCF(8,8)/6-31G(d) level are 5 and 13 kcal/mol. Usually,
the lowest singlet state of carbenes is closed-shell, but the
different energy ordering for 18 can be understood by noting
that the open-shell singlet S1 can acquire aromaticity by moving
a p electron from the carbene center to the oxygen (Scheme
11). This is not possible for the closed-shell S2 singlet.
Scheme 9. Absence of Benzoxirene 23
5. Calculations. To shed some light on the observed
rearrangements in Schemes 4 and 7, we performed DFT
calculations with the relatively new and reliable27 M06-2X
hybrid functional together with a triple-ξ quality basis set (6-
311++G(d,p)). The computed free energies (relative to 1) are
given in Scheme 10.
In agreement with previous studies,10 benzoxetone (2) is
formed with relative ease from ketene 1, requiring only a free
energy of activation of ∼14 kcal/mol. Bisketene 17, which is
formed by electrocyclic ring opening, lies 44 kcal/mol above
the cyclic ketene 1, and an activation barrier of 47 kcal/mol is
Scheme 11. Resonance Structures of the Open-Shell Singlet
Carbene S1-18
Scheme 10. M06-2X/6-311++G(d,p)-Calculated Free
Energies (Numbers in Parentheses) of Ground and
a
Transition States (kcal/mol) Relative to 1
Similarly to diazoketone 1420d and 2-diazo-1-naphthoqui-
none,29 ketene 1 may undergo CO elimination and Wolff
rearrangement to 13 via carbene 18 or in a concerted manner.
The concerted singlet-state reaction has a calculated barrier of
64 kcal/mol (Scheme 10); the stepwise carbene reaction is
expected to be competitive: the S1 state of 18 lies about 60
kcal/mol above 1, and the Wolff rearrangement barrier29b is
about 7 kcal/mol.
The ring-contracted ketene 13 + CO are more stable than 18
by 47 kcal/mol at the M06-2X computational level. Quenching
of 13 leads to the methyl cyclopentadienecarboxylates 19−21.
The initially formed but unobserved 2,4-cyclopentadiene
isomer 19 is the least stable of the three (but only by a few
kcal/mol), and it interconverts with the methyl 1,3-cyclo-
pentadiene-1-carboxylate (20) with a barrier of 21 kcal/mol.
The subsequent 1,5-H shift to give 21 requires 24 kcal/mol.
The energetics of the three isomers and their interconversions
are in good agreement with the finding that the major product
20 is also the most stable. The rearrangement to the minor
product 21 is slightly more feasible than the conversion to the
least stable 19. The trapping product of 1 with MeOH, methyl
salicylate 7, is the global minimum in the investigated sequence,
mostly due to its aromatic character, with a free energy of −41
kcal/mol relative to 1.
The observed labeling of 7b (Scheme 7) and the
interpretation given in Scheme 7 are confirmed by the
calculations. Ketene 1a can rearrange to 1b, giving the labeling
pattern shown in 7b, before it loses CO to form 13. The latter
process has a barrier of ca. 64 kcal/mol, which is significantly
higher than the energies required for the rearrangements
interconverting 1a and 1b.
The transition state for the formation of the unobserved
benzoxirene 23 is predicted to lie at 83 kcal/mol relative to 1,
i.e., well above the transition state for the ring contraction
reaction 1 → 13 (Schemes 9 and 10). Thus, the formation of
a
The T0−S1 (open-shell singlet) energy splitting in carbene 18 is from
a CASSCF(8,8)/6-311++G(d,p) calculation.
6983
dx.doi.org/10.1021/jo5011087 | J. Org. Chem. 2014, 79, 6978−6986

The Journal of Organic Chemistry
Article
(a) A solution of thallium trifluoroacetate30 was prepared by adding
50 g of Tl(III) oxide to 200 mL of TFA in a 500 mL flask. The
suspension was vigorously stirred and refluxed for 12 h under
exclusion of light. The solution thus obtained was filtered to yield a
0.88 M solution of Tl trifluoroacetate, which was then diluted to 0.69
M with trifluoroacetic acid. (b) A 5.9 mL portion of this solution was
added to 500.5 mg (4.07 mmol) of benzoic-1-13C acid (95 atom %
13C). The mixture was stirred magnetically at 73 °C for 24 h and then
cooled to rt. A 5 mL portion of 2.3 N aqueous KI solution was added,
and the mixture was stirred for 15 min. Sodium thiosulfate (250 mg)
was added, and stirring was continued for another 15 min. The
resulting solution was cooled in ice, made alkaline with 4 N NaOH,
filtered, reacidified, and extracted with diethyl ether. The extract was
concentrated and purified by column chromatography on silica gel,
eluting with CHCl3−MeOH 97:3 (Rf = 0.3) to yield 568 mg (57%) o-
iodobenzoic-1-13C acid, mp 162 °C (lit.31 for the unlabeled material
162 °C). (c) The foregoing compound (565 mg; 2.26 mmol) was
mixed with 2.3 mL of 1 N NaOH and heated at reflux for 5 min.
Sodium acetate (900 mg), Cu(II) acetate (74 mg), and water (7 mL)
were then added, and the resulting mixture was refluxed for 1 h. After
cooling, the solution was filtered and the filtrate acidified. The product
was extracted with ether and purified by column chromatography on
silica gel, eluting with CHCl3−MeOH 9:1, thus affording 245 mg
(78%) of a white powder with an Rf value identical to that of salicylic
acid and mp 158.5−159 °C (159 °C for the unlabeled material). (d)
This sample (245 mg; 1.77 mmol) of salicylic-1-13C acid was dissolved
in 5 mL of anhydrous ether, and 3.6 mL (1.98 mmol) of a previously
titrated 0.55 M solution of diazomethane in ether was added at 0 °C.
This mixture was stirred magnetically at 0 °C for 5 min. The ether was
then removed in vacuo at 0 °C, leaving 260 mg (97%) of the desired
ester, having Rf identical with that of unlabeled methyl salicylate
benzoxirene 23 is not intrinsically impossible, but its formation
is not going to be likely under any reaction conditions, since its
ring opening and Wolff rearrangement will always be very facile
reactions.
CONCLUSION
■
The α-oxoketene 1 is formed by FVT of methyl salicylate 7 and
characterized by matrix-isolation IR spectroscopy as well as
online mass spectrometry. Compound 1 is also formed by FVT
of 2-phenylbenzo-1,3-dioxan-4-one (8), phthalic peranhydride
(9), and benzofuran-2,3-dione (11) and by matrix photolysis of
9, 11, and 2-diazocyclohepta-4,6-dien-1,3-dione (12). α-
Oxoketene (1) undergoes elimination of CO at FVT
temperatures above 600 °C with Wolff rearrangement to
fulven-6-one (13) either concertedly or via the α-oxocarbene
18. FVT of methyl 1-13C-salicylate 1a revealed a complex
rearrangement by electrocyclic ring opening to bisketene 17a,
1,5-H shift to 17b, recyclization to 1b, and Wolff rearrangement
to 13a. The resulting ketenes 1a/1b and 13a were trapped with
MeOH to generate the corresponding methyl esters 7b, 20, and
21 (Scheme 7). The 13C labeling results demonstrate that
benzoxirene 23 is not involved.
M06-2X/6-311++G(d,p) calculations confirm the feasibility
of the reaction mechanism. The calculated activation energies
for the reactions of α-oxoketene 1 and the ring-opened
bisketene 17 allow the preferential formation of the major
product 7 before ketene 13 is formed in a Wolff-type
rearrangement. CASSCF calculations indicate that the lowest
singlet state of 6-oxacyclohexa-2,4-dienylidene (18) is open-
shell. The reaction sequence (Scheme 10) confirms the
observed 13C labeling pattern in the trapped products and
the absence of formation of benzoxirene 23.
1
prepared in the same manner: H NMR (CDCl3) δ 8.3 (br s, 1H),
7.82 (d, 1H), 7.44 (t, 1H), 6.97 (d, 1H), 6.86 (t, 1H), 3.93 (s, 3H)
(see Figure S6, Supporting Information); 13C NMR (CDCl3) δ 170.6,
161.5, 135.6, 129.8, 119.1, 117.5, 112.3 (13C-labeled carbon, C1), 52.2
(see Figure S7, Supporting Information). For the assignment of peaks,
see the literature.32
EXPERIMENTAL SECTION
FVT-Matrix Isolation of Methyl Salicylate (7). This compound
was introduced into the FVT apparatus from a sample reservoir held at
0 °C with Ar passing over the sample at a pressure of 10−4 hPa.
Pyrolysis started at ∼300 °C with a very low degree of conversion.
Conversion was complete at temperatures of 750 °C and above.
Carbonylcyclohexadienone (1) was best observed at temperatures of
600−700 °C. Fulven-6-one (13) is formed above 700 °C and became
the main product at and above 800 °C.
IR spectrum of 7 (Ar, 10 K): 3372 m, 3040 w, 2966 w, 1750 m,
1625 vs, 1610 vs, 1446 m, 1360 w, 1350 w, 1330 m, 1310 s, 1257 m,
1217 s, 1197 m, 1160 m, 1094 w, 1034 w, 850 w.
IR spectrum of 1 obtained at 600 °C (Ar, 10 K): 2145 s, 2134 vs,
2114 w, 1655 m, 1600 w, 1521 m, 1420 w, 1292 vw, 1262 vw, 1253 vw,
835 w, 806 w cm−1. When ketene 1 is deposited as a neat substance, it
survives heating until 150 K.
IR spectrum of 13 obtained at 800 °C (Ar, 10 K): 2135 vvs, 2131
vs, 1624 vw, 1594 vw, 1457 m, 1403 m, 1327 m, 1076 w, 1072 vw, 896
m cm−1. When ketene 13 is deposited as a neat substance, it survives
heating until 180 K.
Matrix Photolysis of 1. Photochemical conversion of 1 to 2 was
achieved by photolysis of the matrix-isolated 1 at λ> 340 nm, and this
process was reversed on photolysis of 2 at λ>310 nm. Cycling between
1 and 2 was performed at least 6 times without any significant
differences between initial and final spectra.
■
General Methods. The apparatus used for FVT-matrix isolation
and FVT-MS has been described in detail.22 For FVT-matrix isolation,
in order to avoid unwanted formation of CO by back-streaming of
gaseous pyrolysis products onto the hot filament, an externally heated
quartz tube (10 × 0.9 cm i.d.) was flanged directly to the cryostat head,
the so-called “brick-pyrolyzer”.22 Preparative FVT was usually
performed in vacuo of ∼10−3 hPa in 32 × 2 cm i.d. electrically
heated quartz tubes. Two methods were used for trapping of ketenes
with methanol: (i) in “cold trapping” a layer of MeOH is condensed
on a coldfinger cooled in liquid N2, the pyrolysis product is then
condensed on top of it, and at the end a fresh layer of MeOH is
deposited, so that the pyrolysis product is sandwiched between
methanol layers. After thawing, excess MeOH is removed in vacuo,
and the resulting methyl 1,3-cyclopentadienecarboxylates are dissolved
in cold CDCl3 for immediate low-temperature NMR spectroscopy or
allowed to dimerize at room temperature. (ii) In “hot trapping” a
stream of MeOH vapor in N2 gas is passed through a concentric tube
through the FVT tube, terminating 5 cm from the exit of the FVT
tube, so that the gaseous pyrolysis product will mix with the MeOH
vapor before isolation on the liquid N2-cooled coldfinger. In this case,
a vacuum of 0.5−0.8 hPa is maintained during the pyrolysis. Drawings
and photographs of the apparatus have been published.22 Matrix
photolyses were performed using a 1000 W high-pressure Hg/Xe lamp
equipped with a monochromator, appropriate cutoff filters, and a water
filter to eliminate infrared irradiation, or a 75 W low-pressure Hg lamp
(254 nm).
IR spectrum of 2 (Ar, 10 K): 1916 m, 1904 s, 1624 w, 1604 m, 1443
vw, 1340 w, 1173 w, 1162 w, 892 m, 977 w, 824 m.
Photolysis of the matrix-isolated 1 at 254 nm for 2 h resulted in
decreased absorbances due to 1 and increased absorbances ascribed to
fulvene-6-one 13 and CO.
FVT-Matrix Isolation and Photolysis of 2-Phenylbenzodiox-
in-4-one (8). Compound 8 was sublimed into the FVT apparatus at
50 °C with Ar passing over the sample. Pyrolysis commenced at ca.
Methyl Salicylate-1-13C (7a). The following procedures for
preparation of 13C-labeled methyl salicylate were optimized in several
trial runs with unlabeled materials, and the identities of the products
were ascertained by comparison with authentic samples. Warning:
Thallium derivatives are highly poisonous.
6984
dx.doi.org/10.1021/jo5011087 | J. Org. Chem. 2014, 79, 6978−6986

The Journal of Organic Chemistry
Article
400 °C with formation of ketene 1 and benzaldehyde, and above 600
°C ketene 13 was also formed (IR spectra as above).
Photolysis of matrix-isolated 8 at λ>300 nm for 1 h caused nearly
complete conversion to ketene 1 (see Figure S1, Supporting
Information). Subsequent irradiation at λ > 340 nm resulted in the
usual interconversion of 1 and 2.
FVT-Matrix Isolation and Photolysis of Phthalic Peranhy-
dride 9. CAUTION: This compound is highly explosive.33
Peroxide 9 was sublimed into the FVT tube at 25 °C with Ar
passing over the sample. Pyrolysis commenced at an oven temperature
of 450 °C with formation of ketene 1 (IR spectrum as above), and the
additional formation of ketene 13 took place at temperatures above
650 °C.
IR spectrum of 9 (Ar, 10 K): 1780 vs, 1760 m, 1620 m, 1604 m,
1477 m, 1312 s, 1301 vs, 1246 m, 1108 w, 1102 m, 1048 s, 912 w, 853
w cm−1.
Photolysis of the matrix-isolated 9 at λ > 310 nm afforded ketene 1.
At λ > 345 nm, a mixture of 1 and 2 was obtained, and it was possible
to cycle between these two substances (IR spectra as above).
FVT-Matrix Isolation and Photolysis of Benzofurandione
(11). Compound 11 was sublimed into the FVT apparatus in a stream
of Ar at RT. Elimination of CO and formation of ketene 1 was
observable from FVT temperatures of ca. 400 °C. At temperatures of
700 °C and above, ketene 13 was also formed. Compound 2 was not
observed as a product of FVT.
A matrix containing 1 and 13 was obtained by FVT at 800 °C and
then photolyzed at λ > 340 nm. This resulted in the familiar
conversion of 1 to 2, but ketene 13 remained unchanged.
Benzofurandione (11) was also deposited with Ar without pyrolysis:
IR (Ar, 10 K) 1848 m, 1834 m, 1757 s, 1624 m, 1480 w, 1667 m, 1323
m, 1246 w, 1156 vw, 1026 s, 1037 w, 876 m cm−1.
Photolysis of this matrix at λ = 310 10 or λ > 310 nm resulted in
the formation of CO and ketene 1 (IR spectrum as above). Photolysis
λ > 345 or at 380 10 nm caused conversion of 1 to 2 as described
above. Broadband UV−vis photolysis caused the formation of CO,
ketene 1, and a small amount of ketene 13 (IR spectrum as above).
Matrix Isolation and Photolysis of Diazocycloheptadiene-
dione (12). Compound 1234 was sublimed at rt and deposited with
Ar: IR (Ar, 10 K) 2175 s, 1703 w, 1692 vs, 1437 w, 1347 w, 1323 s,
1223 m, 1217 w, 1061 w, 809 m cm−1. Photolysis of the matrix at λ >
295 nm resulted in the elimination of N2 and formation of ketene 1
(IR as above). Photolysis above 360 nm resulted in the usual mixture
of 1 and 2 (see Figure 1 and Figure S10, Supporting Information).
Further photolysis at >300 nm resulted in reversion of 2 to 1.
Photolysis at 254 nm caused slow loss of CO and formation of ketene
13.
those of authentic materials, and peak assignments for 7,32,35 15, and
1636 were made in accord with the literature.
FVT of Methyl Salicylate-1-13C (7a). Trapping with MeOH.
Formation of 7b and Methyl Cyclopentadienecarboxylates 20
and 21. An 80 mg sample of methyl salicylate-1-13C was pyrolyzed as
above at 800 °C, and the products were trapped on the coldfinger with
MeOH (“cold trapping”). After workup as above, NMR spectra were
recorded at −40 °C (see Figure S9, Supporting Information). The
ratio of products 7b:20:21 was 7:4.5:1. 13C NMR of 7 (unlabeled) and
7b (labeled) (CDCl3) δ (integrals for unlabeled/13C labeled
compound): 170.6 (0.47/0), 161.9 (0.68/0), 135.8 (1.09/1.11),
129.9 (1.17/1.15), 119.2 (1.09/1.11), 117.4 (C3, 0.14/51.48), 112.0
(C1, 0.56/25.56), 52.5 (1/1).
13C NMR of 15 (unlabeled) and 20 (labeled) (CDCl3) δ (integrals
for unlabeled/13C labeled compound): 161.9 (0.1/0), 143.1 (C2, 1.0/
25.0), 140.2 (C3, 0.8/0.9), 139.9 (C1, 0.6/28.3), 132.2 (C4, 0.8/0.8),
51.6 (CH3, 1/1), 41.3 (C5, 1.3/28.6). Peak assignments were made in
accord with the literature.32,35,36
13C NMR of 16 (unlabeled) and 21 (labeled) (CDCl3) δ (integrals
for unlabeled/13C labeled compound): 164.7 (0.1/0), 143.2 (C1, 0.7/
20.4), 138.4 (C2, 1.3/22.7), 134.3 (C4, 1.1/0), 130.4 (C3, 1.3/22.7),
51.8 (CH3, 1/1), 42.7 (C5, 1.2/0.5). Peak assignments were made in
accord with the literature.32,35,36
Dimethyl Dicyclopentadienedicarboxylate (22). The product
of FVT of 100 mg of 7a at 800 °C was trapped with MeOH in the gas
phase (“hot trapping”, see the General Methods). After removal of
excess MeOH in vacuo, the residue was purified by chromatography
on silica gel, eluting with CHCl3. A trace of naphthalene was detected
but not examined further. The main fraction was the dimer 22,
identical with the authentic (unlabeled) dimer25 except for the
presence of 13C-labeled carbon atoms: 13C NMR (CDCl3) δ 165.1,
164.9, 146.9, 142.2, 138.8, 137.7, 54.0, 51.0, 50.9, 50.5, 46.9, 46.3, 40.7,
32.7 (see Figure S12, Supporting Information). For peak assignments,
see ref 25b.
Computational Details. Calculations were performed with the
program package Gaussian 09.37 Structures are optimized using the
global hybrid functional M06-2X38 with the 6-311++G(d,p)39 basis
set. The nature of all stationary points as true minima or as first-order
transition states was confirmed by calculating harmonic frequencies.
Gibbs free energies are obtained at 298.15 K under inclusion of zero-
point vibrational energy corrections.40 The wave function stability of
selected transition states and their open-shell character has been
examined; however, no instability or diradical character could be
found. Vibrational spectra are simulated using the B3LYP/6-31G(d)
level of theory41 and scaled by 0.9614 to correct for anharmonicity.
The CASSCF(8,8)/6-31G(d) and CASSCF(8,8)/6-311++G(d,p)
calculations42 on 18 used an active space consisting of 8 electrons
and 8 orbitals.
FVT of Methyl Salicylate (7). Trapping with MeOH.
Formation of Methyl Cyclopentadienecarboxylates 15 and
16. A 100 mg sample of methyl salicylate (7) was distilled into the
pyrolysis tube of the preparative FVT apparatus at rt and pyrolyzed at
800 °C. The pyrolysis products were trapped with MeOH using either
the “hot trapping” or “cold trapping” method (see the General
Methods). After the end of the pyrolysis, the product−methanol
mixture was allowed to warm to −60 °C, and excess MeOH was
removed by bulb-to-bulb distillation in vacuo at −60 to −40 °C. The
sample was then cooled to −70 °C, and ice-cold CDCl3 was injected
to dissolve the sample, which was then transferred rapidly to an NMR
ASSOCIATED CONTENT
■
S
* Supporting Information
Additional IR and NMR spectra (Figures S1−S12) and
calculated structures of carbene 18 (Figure S13). Absolute
energies and Cartesian coordinates for all calculated com-
pounds and imaginary frequencies for transition states. This
material is available free of charge via the Internet at http://
pubs.acs.org.
1
tube at −78 °C. H and 13C NMR spectra were recorded at −40 °C
(see Figure S8, Supporting Information). For methyl salicylate (7)
1
thus reformed: H NMR (CDCl3) δ 10.89 (s, 1H), 7.9−7.3 (m, 4H),
3.93 (s, 3H), 13C NMR (CDCl3) δ 170.6 (COO), 161.9 (C2), 135.8
(C4), 129.9 (C6), 119.2 (C5), 117.4 (C3), 112.0 (C1), 52.5 (CH3).34
1
Methyl 1,3-cyclopentadiene-1-carboxylate (15): H NMR (CDCl3) δ
AUTHOR INFORMATION
■
6.98−6.83 (m, 1H), 6.70 (m, 1H), 6.57 (m, 1H), 3.76 (s, 3H), 3.28 (s,
2H); 13C NMR (CDCl3) δ 161.9 (COO), 143.1 (C2), 140.2 (C3),
139.9 (C1), 132.2 (C4), 51.6 (CH3), 41.3 (C5). Methyl 1,3-
Corresponding Authors
*E-mail: rainer.koch@uni-oldenburg.de.
*E-mail: wentrup@uq.edu.au.
1
cyclopentadiene-2-carboxylate (16): H NMR (CDCl3) δ 7.30 (br s,
1H), 6.76 (m, 1H), 6.46 (m, 1H), 3.82 (s, 3H), 3.18 (s, 2H); 13C
NMR (CDCl3) δ 164.7 (COO), 143.2 (C1), 138.4 (C2), 134.3 (C4),
130.4 (C3), 42.7 (C5), 51.8 (CH3). The spectra were identical with
Notes
The authors declare no competing financial interest.
6985
dx.doi.org/10.1021/jo5011087 | J. Org. Chem. 2014, 79, 6978−6986

The Journal of Organic Chemistry
Article
(22) For designs and descriptions of the FVT-IR, FVT-MS, and
preparative FVT apparatus, see: Wentrup, C. Aust. J. Chem. 2014,
DOI: org/10.1071/CH14096.
(23) de Carvalho, P. S.; Nachtigall, F. M.; Eberlin, M. N.; Moraes, L.
A. B. J. Org. Chem. 2007, 72, 5986.
(24) Wentrup, C.; Lorencak, P. J. Am. Chem. Soc. 1988, 110, 1880.
(25) (a) Peters, D. J. Chem. Soc. 1959, 1761. (b) Minter, D. E.;
Marchand, A. P.; Lu, S.-p. Magn. Reson. Chem. 1990, 28, 623.
ACKNOWLEDGMENTS
■
This work was supported by the Australian Research Council,
The University of Queensland, and the Center for Scientific
Computation at the Universitat Oldenburg. We are indebted to
Messrs Celestin Thetaz and Glen Kelleher for initial synthetic
́ ́
and spectroscopic work.
̈
́
(26) Thetaz, C.; Wentrup, C. J. Am. Chem. Soc. 1976, 98, 1258.
REFERENCES
■
(27) (a) Rzepa, H. S.; Wentrup, C. J. Org. Chem. 2013, 78, 7565.
(b) Koch, R.; Wentrup, C. Eur. J. Org. Chem. 2013, 7914. (c) Koch, R.;
Berstermann, H. M.; Wentrup, C. J. Org. Chem. 2014, 79, 65.
(d) Swart, M.; Sola, M.; Bickelhaupt, F. M. J. Chem. Theory Comput.
2010, 6, 3145. (e) Cohen, A. J.; Mori-Sanchez, P.; Yang, W. Chem. Rev.
2012, 112, 289.
(1) (a) Kollenz, G.; Ebner, S. Acylketenes. In Science of Synthesis;
Thieme: Stuttgart, Germany, 2006; Vol. 23/9, p 271. (b) Tidwell, T.
T. Ketenes, 2nd ed.; Wiley Interscience: J, 2006. (c) Hyatt, A.;
Raynolds, P. W. Org. React. 1994, 45, 159−646.
(2) (a) Maier, H.; Zeller, K. Angew. Chem., Int. Ed. Engl. 1975, 14, 32.
(b) Kirmse, W. Eur. J. Org. Chem. 2002, 2193.
(3) (a) Qiao, G. G.; Meutermans, W.; Wong, M. W.; Traubel, M.;
Wentrup, C. J. Am. Chem. Soc. 1996, 118, 3852. (b) Nguyen, M. T.;
Hajnal, M. R.; Vanquickenborne, L. G. J. Chem. Soc., Perkin Trans. 2
1994, 169. (c) Schulz, T.; Farber, C.; Leibold, M.; Bruhn, C.;
(28) Ertelt, M.; Hrovat, D. A.; Borden, W. T.; Sander, W. Chem.
̈
Eur. J. 2014, 20, 4713.
(29) (a) Li, Q.; Migani, A.; Blancafort, L. J. Phys. Chem. Lett. 2012, 3,
1056. (b) Cui, G.; Thiel, W. Angew. Chem., Int. Ed. Engl. 2013, 52, 433.
(30) McKillop, A.; Hunt, J. D.; Zelesko, M. J.; Fowler, J. S.; Taylor, E.
C.; McGillivray, G.; Kienzle, F. J. Am. Chem. Soc. 1971, 93, 4841.
(31) Rule, H. G.; Hay, W.; Numbers, A. H.; Paterson, T. R. J. Chem.
Soc. 1928, 178.
(32) Matiello, D. L.; Freeman, R. J. Magn. Reson. 1998, 135, 514.
(33) Jones, M., Jr.; Decamp, M. R. J. Org. Chem. 1971, 36, 1536.
(34) Oda, M.; Kasai, M.; Kitahara, Y. Chem. Lett. 1977, 307.
(35) (a) Shapiro, D. L.; Mohrmann, L. E. J. Phys. Chem. Ref. Data
1977, 6, 919. (b) Mazza, D. Del; Reinecke, M. G.; Smith, W. B. Org.
Magn. Reson. 1980, 14, 540.
̈
Baumann, W.; Selent, D.; Porsch, T.; Holthausen, M. C.; Siemeling, U.
Chem. Commun. 2013, 49, 6834.
(4) Wentrup, C.; Finnerty, J. J.; Koch, R. Curr. Org. Chem. 2010, 14,
1586.
(5) (a) Visser, P.; Zuhse, R.; Wong, M. W.; Wentrup, C. J. Am. Chem.
Soc. 1996, 118, 12598. (b) Leung-Toung, R.; Wentrup, C. Tetrahedron
1992, 48, 7641.
(6) Gompper, R.; Seibold, G.; Schmolke, B. Angew. Chem., Int. Ed.
Engl. 1968, 7, 389.
(7) Buxton, P. C.; Fensome, M.; Heaney, H.; Mason, K. G.
Tetrahedron 1995, 51, 2959.
(8) Chapman, O. L.; McIntosh, C. L. J. Am. Chem. Soc. 1970, 92,
7001.
(36) (a) Grunewald, G. L.; Davis, D. P. J. Org. Chem. 1978, 43, 3074.
(b) Top, S.; Lehn, J.-S.; Morel, P.; Jaouen, G. J. Organomet. Chem.
1999, 583, 63.
(37) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,
N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
(9) (a) Chapman, O. L.; McIntosh, C. L.; Pacansky, J.; Calder, G. V.;
Orr, G. J. Am. Chem. Soc. 1973, 95, 4061. (b) Chapman, O. L.;
McIntosh, C. L.; Pacansky, J.; Calder, G. V.; Orr, G. J. Am. Chem. Soc.
1973, 95, 6135.
(10) (a) Liu, R. C.-Y.; Lusztyk, J.; McAllister, M.; Tidwell, T. T.;
Wagner, B. D. J. Am. Chem. Soc. 1998, 120, 6247. (b) Um, I.-H; Seo, J.-
A.; Mishima, M. Chem.Eur. J. 2011, 17, 3021.
(11) (a) Ziegler, E.; Sterk, H. Monatsh. Chem. 1968, 99, 1958.
(b) Ziegler, E.; Kollenz, G. Monatsh. Chem. 1970, 101, 97.
(12) Kamat, S. P.; Paknikar, S. K. Indian J. Chem., Sect. B 1985, 24B,
38.
(13) (a) Horspool, W. M.; Khandelwal, G. P. J. Chem. Soc. C 1971,
3328. For further photolyses of 11, see: (b) Aycard, J.-P.; Volanschi,
E.; Hnach, M.; Zineddine, H.; N’Guessan, T. Y. J. Chem. Res., Synop.
1995, 346. (c) Aycard, J.-P.; Volanschi, E.; Hnach, M.; Zineddine, H.;
N’Guessan, T. Y. J. Chem. Res., Miniprint 1995, 2068. (d) Itoh, T.;
Tatsugi, J.; Tomioka, H. Bull. Chem. Soc. Jpn. 2009, 82, 475.
(14) Dvorak, V.; Kolc, J.; Michl, J. Tetrahedron Lett. 1972, 13, 3443.
(15) Wentrup, C.; Heilmayer, W.; Kollenz, G. Synthesis 1994, 1219.
(16) See also: Keeffe, J. R. J. Phys. Org. Chem. 2004, 17, 1075.
(17) Schulz, R.; Schweig, A. Tetrahedron Lett. 1979, 20, 59.
(18) (a) Mamer, O. A.; Rutherford, K. G.; Seidewande, R. J. Can. J.
Chem. 1974, 52, 1983. (b) Mamer, O. A. Can. J. Chem. 1972, 50, 2513.
Mamer, O. A. Can. J. Chem. 1971, 49, 3602.
(19) (a) Kappe, C. O.; Wong, M. W.; Wentrup, C. J. Org. Chem.
1995, 60, 1686. (b) Zuhse, R. H.; Wong, M. W.; Wentrup, C. J. Phys.
Chem. 1996, 100, 3917.
(20) Fulven-6-one (carbonylcyclopentadiene): (a) 2133 (vs), 2130
(s) cm−1 in CO matrix: Baird, M. S.; Dunkin, I.; Hacker, N.; Poliakoff,
M.; Turner, J. J. J. Am. Chem. Soc. 1981, 103, 5190. (b) 2133, 2119
cm−1 neat at 77 K: Bloch, R. Tetrahedron Lett. 1978, 1071. (c) 2138
(bs), 2080 (m) in Ar matrix: Schulz, R.; Schweig, A. Angew. Chem., Int.
Ed. Engl. 1984, 23, 509. (d) 2130 cm−1 in hexane: Burdzinski, G.;
̈
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian, Inc.,
Wallingford, CT, 2009.
(38) Zhao, Y.; Truhlar, D. G. Theor. Chem. Acc. 2008, 120, 215.
(39) (a) McLean, A. D.; Chandler, G. S. J. Chem. Phys. 1980, 72,
5639. (b) Krishnan, R.; Binkley, J. S.; Seeger, R.; Pople, J. A. J. Chem.
Phys. 1980, 72, 650. (c) Frisch, M. J.; Pople, J. A.; Binkley, J. S. J.
Chem. Phys. 1984, 80, 3265.
(40) Fabri, C.; Szidarovszky, T.; Magyarfalvi, G.; Tarczay, G. J. Phys.
Chem. A 2011, 115, 4640.
(41) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (b) Lee, C.;
Yang, W.; Parr, R. G. Phys. Rev. B: Condens. Matter 1988, 37, 785.
(42) (a) Roos, B. O.; Taylor, P. R.; Siegbahn, P. E. M. Chem. Phys.
1980, 48, 157. (b) Siegbahn, P. E. M.; Almlof, J.; Heiberg, A.; Roos, B.
̈
O. J. Chem. Phys. 1981, 74, 2384. (c) Yamaguchi, Y.; Frisch, M. J.;
Gaw, J.; Schaefer, H. F., III; Binkley, J. S. J. Chem. Phys. 1986, 84, 2262.
́
Kubicki, J.; Sliwa, M.; Rehault, J.; Zhang, Y.; Vyas, S.; Luk, H. L.;
Hadad, C. M.; Platz, M. S. J. Org. Chem. 2013, 78, 2026.
(21) Wentrup, C.; Gross, G. Angew. Chem., Int. Ed. Engl. 1983, 22,
543.
6986
dx.doi.org/10.1021/jo5011087 | J. Org. Chem. 2014, 79, 6978−6986